Robust Meta Learning for Image based tasks
This addresses the problem of distribution shifts in meta-learning for image tasks, but it appears incremental as it builds on existing meta-learning frameworks.
The paper tackles the challenge of generalization under unknown test distributions by proposing a robust meta-learning method for image-based tasks, demonstrating improved generalization and robustness to distribution shifts.
A machine learning model that generalizes well should obtain low errors on unseen test examples. Thus, if we learn an optimal model in training data, it could have better generalization performance in testing tasks. However, learning such a model is not possible in standard machine learning frameworks as the distribution of the test data is unknown. To tackle this challenge, we propose a novel robust meta-learning method, which is more robust to the image-based testing tasks which is unknown and has distribution shifts with training tasks. Our robust meta-learning method can provide robust optimal models even when data from each distribution are scarce. In experiments, we demonstrate that our algorithm not only has better generalization performance but also robust to different unknown testing tasks.